Resolution assessment of super-resolution microscopy imaging: structural and technical dependencies for cell biology

Super-resolution fluorescence microscopy (SRM) has enabled visualization of nanoscale cellular structures, but systematic evaluation of resolution assessment methods across diverse biological structures and SRM modalities remains lacking. Here, we comparatively assessed three resolution metrics-Full Width at Half Maximum (FWHM), decorrelation analysis, and Fourier Ring Correlation (FRC)-across two SRM techniques (Super-resolution Radial Fluctuation, SRRF; Stimulated Emission Depletion, STED) using key subcellular structures: microtubules (filaments), mitochondria (membranes), and nuclear pore protein Nup98 (single particles) in HeLa/U2OS cells. Our results showed decorrelation analysis provided robust resolution estimates across all structures and modalities (confocal/SRRF/STED), exhibiting superior performance for dense nuclear pore complexes where FWHM failed due to overlapping point spread functions. FWHM was effective for discrete structures (microtubules/mitochondria) but unsuitable for densely packed features. NanoJ-SQUIRREL's integrated FRC analysis successfully quantified SRRF resolution for all structures, with resolution-scaled Pearson coefficient (RSP > 0.90) confirming minimal reconstruction artifacts. STED achieved significantly higher resolution than SRRF, but severe photobleaching prevented FRC application. Our study demonstrates decorrelation analysis offers universal robustness across structures and modalities, FWHM is suitable for discrete structures, and NanoJ-SQUIRREL provides standardized artifact validation and FRC-based resolution quantification for computational super-resolution. This study establishes a framework for context-appropriate resolution assessment in cell biology, emphasizing integration of structural features and technical principles.